Density plots show the distribution of intensities or abundances across samples. Peaks in the plot indicate common values, while the spread indicates variability. Use this plot to identify the range and most frequent values in your data.
Boxplots display the spread and outliers of values for each sample. The box represents the interquartile range (IQR), the line inside the box is the median, and the whiskers extend to 1.5 * IQR from the quartiles. Points outside this range are considered outliers. Use boxplots to compare distributions and identify outliers.
Violin plots combine boxplots and density plots to show the distribution of values. They provide a summary of the data's range, central tendency, and distribution shape. Use violin plots to understand the full distribution and compare between groups.
Mean Time Correlation plots summarize the correlation of each feature with time, highlighting time-dependent trends. Positive correlations indicate increasing trends, negative correlations indicate decreasing trends. Use this plot to identify features that change over time.
Purpose: Lag-1 Differences plots illustrate the changes in feature values between successive time points. What It Shows: This plot helps identify the variability and consistency in the changes of feature values over time. The mean lag-1 difference indicates the average change between time points. The standard deviation of the lag-1 differences indicates the variability in these changes. How to Use It: Use this plot to understand the magnitude and variability of changes in feature values over time. Identify features with consistent increases or decreases and those with high variability between time points.
First Lag Autocorrelation plots illustrate the temporal dependencies within each feature by calculating the autocorrelation of the feature with itself at a lag of one time point. This plot helps identify features that have consistent patterns over time. A high positive autocorrelation indicates that the feature values are similar to their previous values, suggesting a trend or periodicity. A high negative autocorrelation suggests an alternating pattern over time. Use this plot to understand the persistence and cyclic behavior of the features across time points, and to identify features that exhibit stable or periodic patterns.
Purpose: Coefficient of Variation (CV) plots illustrate the relative variability of each feature by calculating the CV for each feature. What It Shows: This plot helps identify features with high or low relative variability. The mean CV indicates the average relative variability across features. The standard deviation of CV indicates the variability in the relative variability across features. How to Use It: Use this plot to understand the consistency and variability of feature values. Identify features with consistently high or low variability relative to their mean.
PCA plots visualize the major trends and patterns in high-dimensional data by reducing it to a few principal components. Points close to each other are similar, they are correlated and clustered. Use PCA plots to identify clustering and variance explained by the principal components.
Scree plots show the amount of variance explained by each principal component. The y-axis represents the percentage of total variance explained, and the x-axis represents the principal components. Use this plot to determine the number of components to consider.
MDS plots display similarities or dissimilarities between samples in a reduced dimension space. Points close to each other are more similar. Use MDS plots to visualize the distance or similarity between samples.
Correlation Heatmaps illustrate the pairwise correlation between all samples. Colors represent the strength of correlation, with a color gradient indicating positive or negative correlations. Use this plot to identify highly correlated samples or groups.
t-SNE plots visualize high-dimensional data by reducing it to two or three dimensions, making complex patterns and relationships easier to see. Points close to each other in the plot represent samples with similar characteristics. Use t-SNE plots to identify clusters and subgroups within your data, revealing how samples relate to each other based on their features. This helps in understanding the structure and similarities within the dataset, especially for discovering hidden patterns and groups.
